- Physical Review Journals
All JournalsPhysics Magazine
Access Provided by Xinjiang University

Physical Review Fluids

Reuse & Permissions

It is not necessary to obtain permission to reuse this article or its components as it is available under the terms of the Creative Commons Attribution 4.0 International license. This license permits unrestricted use, distribution, and reproduction in any medium, provided attribution to the author(s) and the published article's title, journal citation, and DOI are maintained. Please note that some figures may have been included with permission from other third parties. It is your responsibility to obtain the proper permission from the rights holder directly for these figures.

Export citation

Export citation

Choose format for download:

Download Citation
  • Open Access
  • Access by Xinjiang University

Data-driven control of extreme events in turbulent flows through latent space clustering

Youssef Shehata1,2,*, Kevin Schuurman2, Pablo Domínguez Estévez3, and Nguyen Anh Khoa Doan1,2

  • *Contact author: y.shehata26@imperial.ac.uk
PDF

Phys. Rev. Fluids 11, 093902 – Published 10 September, 2026

DOI: https://doi.org/10.1103/jbgr-wsnn

Abstract

Turbulent flows can exhibit extreme events, which are characterized by sudden bursts in the system observables. These events pose significant challenges for prediction and control owing to their intermittent, high-dimensional, and strongly nonlinear nature. In this work we present a predominantly data-driven control framework for the suppression of extreme events in turbulent flows, leveraging reduced-order modeling for nonlinear compression of high-dimensional flow fields and a data-driven clustering algorithm for the identification of precursors to extreme events. A control law is defined in the low-dimensional latent space and is optimized to efficiently mitigate extreme events through actuation values associated with the preidentified clusters. To address issues in the latent space structure that arise from invariance transformations present in many nonlinear systems, leading to an inflated latent space, we employ symmetry-aware autoencoders to establish a more structured and compact latent space. By analyzing high-fidelity simulations of a canonical chaotic flow (Kolmogorov flow), we demonstrate how the resulting closed-loop dynamics exhibit a substantial reduction (up to ca. 99.4%) in the frequency and intensity of extreme events. The framework exhibits scalability to a more chaotic flow regime characterized by higher Reynolds number without encountering numerical instabilities, achieving a 96.9% reduction in the occurrence of extreme events. The ability to incorporate controller limitations, including actuator latency relevant to practical applications of the framework, has also been demonstrated. This study highlights the efficacy of data-driven methods that require little to no prior knowledge of the underlying system dynamics to achieve effective flow control, thereby providing a pathway for general real-time suppression of extreme events in turbulent flows.

View figure in article

Physics Subject Headings (PhySH)

Article Text

References (53)

  1. S. B. Pope, Turbulent Flows (Cambridge University Press, Cambridge, 2000).
  2. E. N. Lorenz, Deterministic nonperiodic flow, J. Atmos. Sci. 20, 130 (1963).
  3. S. H. Strogatz, Nonlinear Dynamics and Chaos (Chapman and Hall/CRC, Boca Raton, FL, 2024).
  4. K. Dysthe, H. E. Krogstad, and P. Müller, Oceanic rogue waves, Annu. Rev. Fluid Mech. 40, 287 (2008).
  5. C. Kharif, E. Pelinovsky, and A. Slunyaev, Rogue Waves in the Ocean (Springer, Berlin, 2009).
  6. D. R. Easterling, J. L. Evans, P. Y. Groisman, T. R. Karl, K. E. Kunkel, and P. Ambenje, Observed variability and trends in extreme climate events: A brief review, Bull. Am. Meteorol. Soc. 81, 417 (2000).
  7. R. B. Alley, J. Marotzke, W. D. Nordhaus, J. T. Overpeck, D. M. Peteet, R. A. Pielke, R. T. Pierrehumbert, P. B. Rhines, T. F. Stocker, L. D. Talley, and J. M. Wallace, Abrupt climate change, Science 299, 2005 (2003).
  8. V. Dakos, M. Scheffer, E. H. van Nes, V. Brovkin, V. Petoukhov, and H. Held, Slowing down as an early warning signal for abrupt climate change, Proc. Natl. Acad. Sci. USA 105, 14308 (2008).
  9. M. Farazmand and T. P. Sapsis, A variational approach to probing extreme events in turbulent dynamical systems, Sci. Adv. 3, e1701533 (2017).
  10. P. J. Blonigan, M. Farazmand, and T. P. Sapsis, Are extreme dissipation events predictable in turbulent fluid flows? Phys. Rev. Fluids 4, 044606 (2019).
  11. T. Lieuwen, V. McDonell, E. Petersen, and D. Santavicca, Fuel flexibility influences on premixed combustor blowout, flashback, autoignition, and stability, J. Eng. Gas Turbines Power 130, 011506 (2008).
  12. M. Hassanaly and V. Raman, Classification and computation of extreme events in turbulent combustion, Prog. Energy Combust. Sci. 87, 100955 (2021).
  13. M. Floris, T. Shiva Sai, D. Nayak, I. Langella, K. Aditya, and N. A. K. Doan, Data-driven identification of precursors of flashback in a lean hydrogen reheat combustor, Proc. Combust. Inst. 40, 105524 (2024).
  14. J. Page, M. P. Brenner, and R. R. Kerswell, Revealing the state space of turbulence using machine learning, Phys. Rev. Fluids 6, 034402 (2021).
  15. R. Maulik, B. Lusch, and P. Balaprakash, Reduced-order modeling of advection-dominated systems with recurrent neural networks and convolutional autoencoders, Phys. Fluids 33, 037106 (2021).
  16. N. A. K. Doan, W. Polifke, and L. Magri, Auto-encoded reservoir computing for turbulence learning, in International Conference on Computational Science (Springer, Cham, 2021), pp. 344–351.
  17. A. J. Linot and M. D. Graham, Data-driven reduced-order modeling of spatiotemporal chaos with neural ordinary differential equations, Chaos 32, 073110 (2022).
  18. C. E. P. De Jesús, Carlos E, A. J. Linot, and M. D. Graham, Building symmetries into data-driven manifold dynamics models for complex flows, Phys. Rev. Fluids 10, 064401 (2025).
  19. T. Kurihana, I. Mastilovic, L. Wang, A. Meray, S. Praveen, Z. Xu, M. Memarzadeh, A. Lavin, and H. Wainwright, Identifying climate patterns using clustering autoencoder techniques, AIES 3 (2024).
  20. F. R. Spuler, M. Kretschmer, M. A. Balmaseda, Y. Kovalchuk, and T. G. Shepherd, Learning Predictable and Informative Dynamical Drivers of Extreme Precipitation using Variational Autoencoders (Copernicus Publications, Göttingen, Germany, 2025).
  21. K. Aoki, H. Nakano, N. Hirose, N. Usui, K. Sakamoto, T. Toyoda, S. Urakawa, and Y. Kawakami, Application of variational autoencoder–based clustering for geophysical fluid circulations with small sample size, J. Geophys. Res.: Mach. Learn. Comput. 3, e2025JH001051 (2026).
  22. N. A. K. Doan, W. Polifke, and L. Magri, Short- and long-term predictions of chaotic flows and extreme events: A physics-constrained reservoir computing approach, Proc. R. Soc. A 477, 20210135 (2021).
  23. A. Racca and L. Magri, Data-driven prediction and control of extreme events in a chaotic flow, Phys. Rev. Fluids 7, 104402 (2022).
  24. U. Golyska and N. A. K. Doan, Clustering-based identification of precursors of extreme events in chaotic systems, in Computational Science–ICCS 2023, Lecture Notes in Computer Science Vol. 10476, edited by J. Mikyška, C. de Mulatier, M. Paszynski, V. V. Krzhizhanovskaya, J. J. Dongarra, and P. M. Sloot (Springer Nature, Cham, 2023), pp. 313–327.
  25. V. Pyragas and K. Pyragas, Using reservoir computer to predict and prevent extreme events, Phys. Lett. A 384, 126591 (2020).
  26. W. E. Sherwood, Fitzhugh–Nagumo model, in Encyclopedia of Computational Neuroscience, edited by D. Jaeger and R. Jung (Springer New York, New York, 2013), pp. 1–11.
  27. A. Racca and L. Magri, Control-aware echo state networks (Ca-ESN) for the suppression of extreme events, in Machine Learning and the Physical Sciences Workshop, arXiv:2308.03095.
  28. M. Farazmand and T. P. Sapsis, Closed-loop adaptive control of extreme events in a turbulent flow, Phys. Rev. E 100, 033110 (2019).
  29. A. M. Obukhov, Kolmogorov flow and laboratory simulation of it, Russ. Math. Surv. 38, 113 (1983).
  30. N. Platt, L. Sirovich, and N. Fitzmaurice, An investigation of chaotic Kolmogorov flows, Phys. Fluids 3, 681 (1991).
  31. D. Kelshaw, KolSol: Pseudospectral Kolmogorov flow solver, Github, 2022, https://github.com/MagriLab/KolSol.
  32. C. Canuto, M. Y. Hussaini, A. Quarteroni, and T. A. Zang, Spectral Methods in Fluid Dynamics (Springer, Berlin, 1988).
  33. J. L. Lumley, The Structure of Inhomogeneous Turbulent Flows, Atmospheric Turbulence and Radio Wave Propagation, edited by A. M. Yaglom and V. I. Tatarsky (Nauka, Moscow, 1967), pp. 166–178.
  34. G. Berkooz, P. Holmes, and J. L. Lumley, The proper orthogonal decomposition in the analysis of turbulent flows, Annu. Rev. Fluid Mech. 25, 539 (1993).
  35. P. J. Schmid, Dynamic mode decomposition of numerical and experimental data, J. Fluid Mech. 656, 5 (2010).
  36. T. Murata, K. Fukami, and K. Fukagata, Nonlinear mode decomposition with convolutional neural networks for fluid dynamics, J. Fluid Mech. 882, A13 (2020).
  37. E. Plaut, From principal subspaces to principal components with linear autoencoders, arXiv:1804.10253v3.
  38. S. Kneer, T. Sayadi, D. Sipp, P. Schmid, and G. Rigas, Symmetry-aware autoencoders: s-PCA and s-nlPCA, arXiv:2111.02893.
  39. K. Zeng and M. D. Graham, Symmetry reduction for deep reinforcement learning active control of chaotic spatiotemporal dynamics, Phys. Rev. E 104, 014210 (2021).
  40. M. Jaderberg, K. Simonyan, A. Zisserman, and K. Kavukcuoglu, Spatial transformer networks, Advances in Neural Information Processing Systems 28 (NIPS, 2015), pp. 2017–2025.
  41. P. J. Schmid, A. García-Gutierrez, and J. Jiménez, Description and detection of burst events in turbulent flows, J. Phys.: Conf. Ser. 1001, 012015 (2018).
  42. P. J. Schmid, O. T. Schmidt, A. Towne, and M. J. P. Hack, Analysis and prediction of rare events in turbulent flows, Advances in Neural Information Processing Systems 28 (NIPS 2015) (Curran Associates, Inc., Red Hook, NY, 2018), pp. 2017–2025.
  43. A. G. Nair, C.-A. Yeh, E. Kaiser, B. R. Noack, S. L. Brunton, and K. Taira, Cluster-based feedback control of turbulent post-stall separated flows, J. Fluid Mech. 875, 345 (2019).
  44. I. Loshchilov and F. Hutter, SGDR: Stochastic gradient descent with warm restarts, in International Conference on Learning Representations (ICLR, 2017).
  45. T. A. Howell, S. Le Cleac'h, S. Singh, P. Florence, Z. Manchester, and V. Sindhwani, Trajectory optimization with optimization-based dynamics, IEEE Rob. Autom. Lett. 7, 6750 (2022).
  46. B. Sukhija, N. Köhler, M. Zamora, S. Zimmermann, S. Curi, A. Krause, and S. Coros, Gradient-based trajectory optimization with learned dynamics, in 2023 IEEE International Conference on Robotics and Automation (ICRA) (IEEE, London, UK, 2022), pp. 1011–1018.
  47. D. P. Kingma and J. Ba, Adam: A method for stochastic optimization, in 3rd International Conference on Learning Representations (ICLR 2015) (OpenReview.net, San Diego, CA, USA, 2014).
  48. Y. Bengio, P. Simard, and P. Frasconi, Learning long-term dependencies with gradient descent is difficult, IEEE Trans. Neural Netw. 5, 157 (1994).
  49. R. Pascanu, T. Mikolov, and Y. Bengio, On the difficulty of training recurrent neural networks, in Proceedings of the 30th International Conference on Machine Learning, Proceedings of Machine Learning Research Vol. 28, edited by S. Dasgupta and D. McAllester (PMLR, Atlanta, 2013), pp. 1310–1318.
  50. A. Cleary and J. Page, Characterizing the Reynolds number dependence of the chaotic attractor in two-dimensional turbulence with dimension-minimizing autoencoders, Phys. Rev. E 112, 055105 (2025).
  51. M. Rudolph, B. Wandt, and B. Rosenhahn, Structuring autoencoders, in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops (IEEE, Seoul, South Korea, 2019), pp. 615–623.
  52. Y. Shehata, K. Schuurman, P. D. Estevez, and N. A. K. Doan, Kolmogorov flow at Reynolds number = 40, Zenodo, 2026, https://doi.org/10.5281/zenodo.21853176.
  53. J. Moehlis, H. Faisst, and B. Eckhardt, A low-dimensional model for turbulent shear flows, New J. Phys. 6, 56 (2004).

Outline

Information

Phys. Rev. Fluids 11, 093902– Published 10 September, 2026
  • Received 25 November 2025
  • Accepted 4 August 2026
Crossmark - Check for updates button
Creative Commons Logo

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article's title, journal citation, and DOI.

Published by the American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation